Strange findings about voting for primary day
In celebration of primary day in Kentucky and Oregon, the Nudge blog offers five strange findings about voting that are not explained by the usual – individually rational – factors driving voter turnout and vote choice, which are income, education, party identification, ideological leaning.
Partisans in enemy territory: James Gimpel, Joshua Dyck, and Daron Shaw find that when it comes to voting, a Republican surrounded by Democratic neighbors does not behave the same as a Democrat surrounded by Republican neighbors. Looking at 2000 data in seven states, including Florida, they find that Republican partisans in Democratic neighborhoods vote less than expected, even after accounting for socioeconomic status. These losses are not offset by Democratic vote depression in heavily Republican neighborhoods.
Rain dampens turnout — for Democrats*: Brad Gomez, Thomas Hansford, and George Krause provide empirical evidence for what political strategists have claimed for years: Republicans should pray for rain in November. Rain reduces voter participation by a rate of almost 1 percent per inch (an inch of snowfall decreases turnout by almost .5 percent. Looking at data from 1960-2000, they find that for every inch increase in rain above its election day normal, the Republican presidential candidate received approximately an extra 2.5% of the vote. For every inch increase in snow above normal, the Republican candidate’s vote share increases by approximately .6 percent. Democratic big whig Donna Brazile, the manager of Al Gore’s campaign in 2000, watched the Weather Channel regularly before the 2006 mid-term elections to see where the storms would be.
The election board is watching: Alan Gerber, Don Green, and Christopher Larimer find that people are more likely to vote if they think their behavior will become publicly known. The trio conducted a field experiment in which they sent letters intended to influence a person’s likelihood of voting. Voter turnout increased 4.5 percent over the control group when households were shown their own voting record. It rose 8.1 percent over the control group when voters were shown their own voter record and the voting records of their neighbors. Nothing like the threat of shaming to get someone to the polling booth! For comparative purposes, voter mobilization campaigns that make personal contact with voters at their front doors and boost turnout by 3 percent are considered a huge success.
The environmental effect of a polling place: Jonah Berger, Marc Meredith, and S. Christian Wheeler say that your polling place exerts a strange influence on your voting decisions. People who vote in schools are more likely to support propositions for more education funding. Berger, Meredith, and Wheeler used 2000 data from an Arizona referendum proposing to raising the state sales tax from 5.0 percent to 5.6 percent with the money going to education. By a count of 55 percent to 53.09 percent, voters in schools were more likely to support this initiative than those in other polling places like churches or community centers.
The Black Democrat boost: Ebonya Washington finds that Black Democrats on the ballot boost voter turnout among blacks and whites. Both groups voter turnout increases 2-3 percentage points with each Black Democrat on the ballot (although in absolute numbers, white turnout is greater because of their proportion in the population). Black Republicans on the ballot do not increase turnout. Whites from both parties come to the polls. However, both White Democrats and White Republicans are less likely to vote for the black candidate. The absence of a similar turnout response to a Black Republican may indicate that perceptions of the black candidate’s ideology may be behind the result.
*For the abstract of a paper that refutes the second finding, click here.